Abstract
This publication describes systems and techniques for evaluating the operational status or “health” of a heating-ventilation-air-conditioning (HVAC) system by analyzing sound generated by the HVAC system. The sound may be monitored by a microphone incorporated in a smart thermostat or other smart home device and automatically evaluated using an HVAC Large Language-Type Model (HVAC-LLM) including acoustic signatures of properly and/or improperly operating HVAC systems. For example, a system includes one or more microphones integrated in a smart thermostat and/or one or more other smart home devices that act as passive sensors to capture sound made by the HVAC system. Initialization of the system includes capturing a baseline sound, converting the captured sound into a spectrogram, tokenizing the spectrogram, and rendering a vectorized embedding representing the sound to create a baseline acoustic signature for use by the HVAC-LLM. Responsive to a user request, a current acoustic signature is generated from current sound captured by the one or more microphones, and the current acoustic signature is compared to the baseline acoustic signature or other acoustic signatures in the HVAC-LLM to determine if the current acoustic signature indicates a potential fault in the HVAC system.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Shin, D, "Smart Device HVAC Health Evaluation with Large-Language Model Acoustic Signature Analysis", Technical Disclosure Commons, (October 27, 2025)
https://www.tdcommons.org/dpubs_series/8791